We present a new approach to reconstructing annual mean temperature using geochemical data from lake sediments. This paper uses Bayesian inference, a type of statistical approach, and creates a model called Simulating Climate Using Bayesian Inference with proxy Data Observations (SCUBIDO), which takes the high-resolution geochemical data and transforms them into quantitative climate information at an annual resolution. We show the results from two lakes in England and Finland to produce temperature reconstructions for the past 8000 years with data every year.